The employment and the field of industrial relations
THE EMPLOYMENT RELATIONSHIP 11THE EMPLOYMENTRELATIONSHIP AND THEFIELD OF INDUSTRIALRELATIONSPAUL EDWARDSThe term ‘industrial relations’ (IR) came into common use in Britain and NorthAmerica during the 1920s. It has been joined by personnel management (PM)and, since the 1980s, human resource management (HRM). All three denote apractical activity (the management of people) and an area of academic enquiry.Texts in all three ﬁelds commonly take as their starting point the corporateassertion that ‘people are our most important asset’: if this is indeed so, there islittle further need to justify a text. Yet we need ﬁrst to explain what lies behindthis apparent axiom. It is then important to highlight some of the key currentissues about the conduct of work in modern Britain. We can then consider howIR as an academic approach addresses these issues and the distinction between itand the other two ﬁelds of enquiry. Finally, the structure of the book is explained. First, some basic explanation. ‘Industry’ is sometimes equated with manufac-turing, as in contrasts between industry and services. ‘Industrial relations’ has inprinciple never been so restricted. In practice, however, attention until recentlyoften focused on certain parts of the economy. These in fact embraced morethan manufacturing to include the public sector for example, but there wasneglect of small ﬁrms and large parts of the private service sector. Whether ornot there were good reasons for this neglect (and the case is at least arguable),the situation has changed, and recent research has addressed growing areas ofthe economy such as call centres. To avoid confusion some writers prefer theterm ‘employment relations’, and if we were starting from scratch this might bethe best label; yet the term ‘industrial relations’ has become sufﬁciently embed-ded that it is retained here to cover relations between manager and worker in allspheres of economic activity. The focus is employment: all forms of economicactivity in which an employee works under the authority of an employer and
2 PAUL EDWARDSreceives a wage in return for his or her labour. Industrial relations thus excludesdomestic labour and also the self-employed and professionals who work on theirown account: the contractual relations between a self-employed plumber andhis customers are not ‘industrial relations’, but the relations between a plumbingﬁrm and its employees are. In the UK self-employment comprises about 12 percent of people in employment (see table 1.1, below). The bulk of the workingpopulation is thus in an employment relationship, with the great majority ofthem, of course, being employees rather than employers. Some writers deﬁne IR no more exactly than the study of all forms of theemployment relationship. This is not sufﬁciently precise to distinguish it fromthe economics or sociology of work. More importantly, there are some distinctemphases in an IR approach which give it a speciﬁc value in explaining theworld of work. These emphases are discussed below. There has been muchdebate over the years as to whether the emphases and analytical preferences ofIR make it a discipline, as distinct from a ﬁeld of study. The view taken in thischapter (which is not necessarily shared by other chapters) is that IR is a ﬁeld ofstudy and not a distinct discipline. Indeed, one of its strengths is its willingnessto draw from different disciplines so that people who specialize in the ﬁeld havedeveloped an analytical approach which is more than the sum total of theapplication of individual disciplines. Even if this view is accepted, there arecompeting views as to the strengths and weaknesses of the approach, and whetherit has responded adequately to the changing nature of work. Some of theseissues are addressed below. Why is paid employment important? It is important to the employee most obvi-ously as a source of income. Note that it is not the case that work outsideemployment is an easy alternative: at one time, it was argued by some that acombination of unemployment, self-provisioning and work in the informaleconomy provided an alternative to the formal economy, but research foundthat such work tends to be additional rather than an alternative to formalemployment. Work is also important to the employee as a means of identity.‘What do you do for a living?’ is a standard query to locate a new acquaintance.And what goes on within the employment relationship is crucial, not only interms of the pay that is earned but also the conditions under which it is earned:the degree of autonomy the employee is granted, the safety of the work envir-onment, the opportunity for training and development, and so on. For the employer the work relationship is crucial in two different senses. First, itis commonly argued that capital and technologies are increasingly readily avail-able, so that a ﬁrm’s competitive position depends on the skills and knowledgeof its workers. Some analytical grounding for this argument comes from theresource-based view of the ﬁrm which developed from debates on strategicmanagement. This view sees the ﬁrm as a bundle of assets and argues that it isthe conﬁguration of these assets, rather than positioning in relation to an exter-nal market, which is central to competitive advantage (Wernerfeld 1984; seefurther chapter 7). Not surprisingly, HRM and IR writers have latched on to thisidea, arguing that ‘distinctive human resources’ are the core resource (Cappelliand Crocker-Hefter 1996). Second, and fundamentally, these ‘human resources’
THE EMPLOYMENT RELATIONSHIP 3Table 1.1 Employment, unemployment and earnings, UKPopulation by employment status (thousands) Total Economically In Unemployed Economically active employment (ILO deﬁnition) inactiveMales aged 16–641990 18,312 16,175 15,027 1,148 2,1362000 19,020 16,034 15,049 984 2,987Females aged 16–591990 16,706 11,912 11,122 790 4,7942000 17,292 12,534 11,916 618 4,758Distribution of employed population (thousands) Employees Self-employed Full-time Part-time Full-time Part-timeAll males1992 10,971 658 2,260 1822000 11,917 1,064 2,029 272All females1992 5,963 4,491 420 3662000 6,489 5,032 427 423Percentage of age group in employment in 2000 Age 18–24 25–34 35–49 50–64 (M) / 50–59 (F) 65+ (M) / 60+ (F)Males 71.3 88.9 88.5 68.8 7.6Females 64.1 71.7 74.9 63.9 8.2Percentage of all unemployed who were out of work > 12 months in2000, by age (ILO deﬁnitions) Age 18–24 25–49 50+Males 20.4 39.3 46.2Females 10.3 21.2 31.6Hourly earnings in £ (all full-time employees) and prices Earnings RPI All Male Female (1987 = 100)1990 6.37 6.88 5.31 126.12000 10.32 11.00 9.02 170.3Figures are for spring each year and seasonally adjusted. Earnings data are derived from theNew Earnings Survey.Source: Labour Market Trends (March 2001).
4 PAUL EDWARDSare different from other resources because they cannot be separated from thepeople in whom they exist. The employment relationship is about organizinghuman resources in the light of the productive aims of the ﬁrm but also the aimsof employees. It is necessarily open-ended, uncertain, and, as argued below, ablend of inherently contradictory principles concerning control and consent. Finally, paid employment is important to society for what it expects in terms of‘inputs’ and produces as ‘outputs’. Inputs include how much labour is demanded(with obvious implications if demand is less than supply, resulting in unemploy-ment) and what types of labour are sought (inﬂuencing, for example, the kindsof skills which ‘society’ provides through the education system). If employmentis structured on gender lines, this will have major consequences for the domesticdivision of labour and the roles of men and women in society; the traditionalimage of the male breadwinner applied not only to paid employment but alsohad implications for the ability of women to engage in politics, the arts, andsport. ‘Outputs’ include not only goods and services but also structures of advant-age and disadvantage. These are properly called structures because they areestablished features of society which are hard to change, for example differencesof pay between occupations and between men and women.Issues in the Regulation of WorkIf work is important, how many people are in an employment relationship andhow many are not, and what has been happening to work relationships? Theexercise in box 1.1 may be helpful. Alongside such trends have been developments in the management of workwhich are analysed in detail in this book. They include a decline in traditionalways in which people represent their views to their employers (termed ‘indirect’or ‘representational’ participation), which in Britain means through a trade union.Associated with a decline of unions has been a reduction in the percentage ofemployees who are covered by collective bargaining. Collective bargaining is akey focus in IR (the term having been coined by two of the UK founders of thesubject, Sidney and Beatrice Webb, at the end of the nineteenth century). Itmeans the negotiation of pay and other conditions of employment between anemployer (or a group of employers) and a trade union acting for its members(see chapter 8). There has also been a growth in ‘direct’ participation, that isinvolvement not through a representative structure but through work-basedactivity; examples are problem-solving groups and teamworking (see chapters 7and 13). The legal framework has also changed rapidly, as discussed in chapters5 and 6. Some of these developments reﬂect developments within Britain itself,some stem from Europe, some from the speciﬁc inﬂuence of multinational com-panies, and some from broader trends in the world economy. These developments in the management of work are highly important inthemselves, in shaping how much autonomy workers have in their work andtheir ability to shape key decisions that affect them. But what goes on within IRcan have substantial effects on wider aspects of society. To take but one example,
THE EMPLOYMENT RELATIONSHIP 5Box 1.1 Labour market participation, pay, and inequality Consider the population of working age. Ofﬁcial statistics distinguish between the economically active and the inactive. The former group is then divided into those currently employed and the unemployed. The employed can be divided according to status (employee or self-employed), whether the work is full- or part-time, and so on. Table 1.1 provides some basic ﬁgures on these categories, together with information on unemployment and earnings. Table 1.2 gives abbreviated data on the distribution of employment by sector. What are the main patterns that can be observed? It may also be useful to consider what such terms as ‘economically active’ mean and how statistics on such things as unemployment and earnings are compiled. There are several important features of work in Britain which need to be borne in mind in considering the implications of the ﬁgures. Details and further discussion can be found in Gregg and Wadsworth (1999). • The number of ‘economically inactive’ people of working age has risen, particularly among the over-ﬁfties. Early retirement is a common means for ﬁrms to shed labour. What might this say about the nature of jobs? • Work has polarized across households: there are more families where all the adults work, and more where no one is in paid employment. What might this say about links between work and home? Regional differences in employment and unem- ployment rates are also substantial, as a glance at the relevant ﬁgures in Labour Market Trends will show. For example, in January 2001 unemployment (based on those claiming unemployment beneﬁt, not the ILO deﬁnition used in table 1.1) was 6.6 per cent in the region of highest unemployment in the UK, the north-east of England, but within that region rates for localities ranged from 3.4 to 12.3 per cent. In the lowest unemployment region, the south-east, the average rate was 1.9 per cent, with localities varying between 0.5 and 7.4 per cent. • If we look at households where someone is in work, the number of hours per week devoted to work has risen since about 1980. Work effort intensiﬁed over the same period, at ﬁrst in manufacturing and then in services (Green 2001). Why might this be, and what does it say about workers’ experience of work? • Men’s employment rates have fallen while women’s have risen. But this latter rise is largely restricted to women with working partners; there is no change for single women or lone parents. The gender pay gap has narrowed, but it remains substan- tial, and has in fact widened for women working part-time. Why? • Wage inequality has risen (as it has in other countries, though generally more slowly) to reach levels higher than during most of the twentieth century. Why?chapter 8 shows that the decline of unions and collective bargaining explainssome of the rise in wage inequality; as mentioned below, moreover, it appearsthat international differences in IR structures help to explain the size of thegender pay gap. It might be helpful to pause to consider what mechanisms mightexplain such links between processes and outcomes, and which of the otherfeatures in the bullet points in box 1.1 could be the result of trends in the handl-ing of IR. IR thus has important implications for life beyond its own terrain. What are the pressing current issues in employment? Three examples aregiven, partly for their substantive importance, but also to signal the critical viewof them which is developing within IR.
6 PAUL EDWARDSTable 1.2 Distribution of employees by sector, UK (thousands) 1990 2000Mining, quarrying, electricity, gas and water 406 204‘Engineering’ 1,544 1,258All other manufacturing 3,334 2,705All services 16,643 18,597 Wholesale and retail trade, repairs 3,741 4,126 Hotels and restaurants 1,207 1,395 Financial intermediation 1,055 992 Renting, research, computers etc 2,410 3,207 Health and social work 2,311 2,541‘Engineering’ = machinery, electrical equipment plus transportation equipment.Source: Summarized from data in Labour Market Trends (March 2001), which contains a moredetailed breakdown. The ﬁrst concerns so-called ‘high-commitment’ or ‘high-involvement’ worksystems. These are discussed in detail in chapter 13, but essentially embracesystems such as teamworking and are often linked to new managerial techni-ques such as Business Process Re-engineering. Some research in the UK andthe US ﬁnds that these systems ‘work’ in that they produce improvements inefﬁciency and (though the evidence is much more controversial here) can beassociated with beneﬁts for workers too (see chapter 19). Yet it is also found thatthey exist only rarely; perhaps 2 per cent of UK workplaces conform to the high-commitment model. This situation is often seen as a paradox. There is some value in posing the matter this way, but there are now somereasonably well-established resolutions of the paradox as posed. As will be seenin chapters 7 and 13 in particular, the beneﬁts depend on certain conditions andthey operate best only in the long term whereas their costs are signiﬁcant andimmediate. The structure of British ﬁrms tends to mean that the conditions arehard to secure and that the short-term considerations outweigh the long-runones. Moreover, what is meant by ‘working’ requires more exploration: workingin what ways and for whom? Other modes of organizing work, notably thosebased on low skills and low wages, can equally work for employers in producingacceptable proﬁts; and, some commentators would argue, they are well suited tothe British context (see chapter 15). And high-commitment systems will havetheir own tensions: they are a way of managing the contradictions of controland consent, not escaping from them. A second pressing issue is the international context. Some writers deployconcepts such as globalization to capture new international competitive pres-sures. They are better seen as convenient labels rather than developed concepts,for issues immediately arise as to the novelty of the developments identiﬁed andwhat identiﬁable social forces are actually causing them. In the ﬁeld of work,
THE EMPLOYMENT RELATIONSHIP 7three interrelated forces are international competition, the role of multinat-ional companies (MNCs), and European integration. Under the ﬁrst, the Britisheconomy has become increasingly open, as indicated by a growth in imports andexports as a proportion of GDP and the use of explicit wage and cost compar-isons by companies in the making of investment decisions (see chapters 3 and7). A well-known UK example is the decision in early 2000 of the German ﬁrmBMW to sell the Rover car company, which it had acquired in 1994, blaming thevalue of the pound in relation to the euro and the difﬁculty of restructuring theRover operations to attain satisfactory productivity levels. That the UK is notalone is illustrated by the case of Renault in Belgium, which in February 1997announced without warning the closure of its Vilvoorde plant with the loss of3,000 jobs. This example also points to one role of the MNC. But, as discussed in chap-ter 4, there are other roles, notably the importing of forms of work organization,and it is often US MNCs which are in the lead here. Finally, European integra-tion has effects through the impact of European labour law on Britain andthrough the wider processes of economic and monetary union (EMU). Underthe ﬁrst, European directives have had clear effects on matters as varied as theregulation of working time, consultation over redundancies and European WorksCouncils (requiring that certain large MNCs establish such councils for the pur-poses of information and consultation about their European operations). Underthe second, unit wage costs are increasingly subject to comparison across Europe,while the implications spread outside the traded goods sector. Thus governmentﬁnances are shaped by pressures on interest rates and the public sector borrow-ing requirement, which in turn has implications for the control of costs, includ-ing pay, in the public sector. One aspect of internationalization which has recently come to the fore iswhether British industrial relations are being Europeanized or Americanized.• Europeanization means either or both of: the inﬂuence of European-level developments in Britain (either directly, for example the application of dir- ectives, or indirectly, for example where monetary union brings pressure for convergence in IR practice); and the development of a common model across Europe. Such a model often embraces ideas of ‘social partnership’. As dis- cussed in chapters 7, 9 and 10, these ideas are often imprecise and contested, but at their core is the notion of a common agenda between representatives of capital and labour.• Americanization embraces the continuing decline of unions and the assertion of a market-driven model.The former process is perhaps the more obvious in the light of European integra-tion and the promulgation of a European social model claiming to combineﬂexibility with security and to promote employee participation without threaten-ing efﬁciency (see Bach and Sisson 2000: 35). As Bach and Sisson stress, how-ever, such a model is a prescription for what might be rather than an account ofwhat exists, and several aspects of it are under challenge from international cost
8 PAUL EDWARDSpressures. At the same time, the rapid growth of the American economy duringthe 1990s and the European interest in its ability to generate jobs indicate thatthe American model of weak trade unions and extensive ﬂexibility is equallyinﬂuential. It is not of course the case that these models are tightly integratedpackages or that one can simply choose between them. Different features can becombined in different ways. A third set of issues concerns ‘outcomes’ of a pattern of IR. The most dis-cussed outcome, touched on above, is economic performance. Chapter 19 discussesthe linkages between IR arrangements and performance. But other outcomesinclude the level and pattern of pay. As indicated above, one of the outstandingfeatures of the British economy has been the rise in income inequality since1980. A closely connected form of outcome is the pattern of gender inequality,as indexed by pay differentials and the degree to which women gain access tothe most desirable occupations (see chapter 16). It has been shown across manyadvanced industrialized countries that various measures of equality and well-being, including the size of the pay gap between men and women and thedegree of pay inequality between the top and bottom of the income distribution,are affected by the extent of collective bargaining (e.g. Whitehouse and Zetlin1999). Given that collective bargaining has been in long-term decline in the UK,key issues are whether this decline is likely to be reversed, and if not what otherarrangements might be put in place and what implications they have for economicwelfare.Analysing the Employment RelationshipComponents of industrial relationsWhat has IR to say about how we might analyse such issues? The employmentrelationship has two parts, market relations and managerial relations (Flanders1974). The former is the more obvious. It covers the price of labour, whichembraces not only the basic wage but also hours of work, holidays and pensionrights. In this respect, labour is like any other commodity, with a price whichrepresents the total cost of enjoying its use. Yet labour differs from all othercommodities in that it is enjoyed in use and is embodied in people. A machine ina factory is also enjoyed in use and for what it can produce. Yet how it is usedis solely up to its owner. The ‘owner’ of labour, the employer, has to persuadethe worker, that is, the person in whom the labour is embodied, to work.Managerial relations are the relationships that deﬁne how this process takesplace: market relations set a price for a set number of hours of work, andmanagerial relations determine how much work is performed in that time, atwhat speciﬁc task or tasks, who has the right to deﬁne the tasks and change aparticular mix of tasks and what penalties will be deployed for any failure tomeet these obligations. A standard text thus deﬁnes IR as the ‘study of the rulesgoverning employment’ (Clegg 1979: 1). The importance of this deﬁnition isdeveloped below.
THE EMPLOYMENT RELATIONSHIP 9 state employee employer employment relationship employee representativesFigure 1.1 The employment relationship The employment relationship is by deﬁnition a relationship between an em-ployee and an employer. As shown in ﬁgure 1.1, this direct relationship may bemediated by the two other key institutions to IR, the trade union (or morerarely a non-union collectivity representing employees) and the state. A tradeunion in its most basic role represents a group of workers in a speciﬁed part oftheir relations with a single employer. A union’s role can be measured in termsof density, extent, mobilization and scope.• Density is the proportion of an identiﬁed constituency who are members of a union.• The extent of a union’s activity refers to the range of the constituency: a union can represent a small group of employees in one locality, or all the employees in an occupation, or all the employees of a given employer, or extend beyond an occupation or an employer.• Mobilization – the degree to which unions identify common interests among their members, persuade the members as to what the interests are, and organize in pursuit of the interests – is important because, most obviously in countries such as France, a union may be capable of mobilizing more employees than its nominal members. By the same token, members will not necessarily follow a union’s policy. Unions face issues of how far they repres- ent members and of aggregating membership interests into a common policy.• Scope is the degree to which the various aspects of the employment relation- ship are within the purview of the union: it may bargain only over wages and hours, or cover also working conditions, or extend further to issues includ- ing training, the classiﬁcation of jobs and the system of workplace discipline. Unions engage with employees through efforts to organize them and throughmobilization around sets of demands. They engage with employers by taking
10 PAUL EDWARDSpart in collective bargaining. They may also engage with the state, for examplein making demands for legislation or in engaging in more lasting forms ofaccommodation (such as ‘corporatism’ in the Nordic countries or a series of‘Accords’ in Australia). The state inﬂuences the employment relationship directly through laws onwages (e.g. minimum wages), working conditions (e.g. on hours of work) andmany other issues, and through its role as the employer of public sector workers(see chapter 11). It also has a series of indirect inﬂuences. It has relationshipswith unions, either through laws on union government, or through bilateralarrangements (e.g. the UK ‘social contract’ of the 1970s in which unions prom-ised to moderate wage demands in return for tax concessions), or throughtrilateral relationships also involving employers (corporatism). In addition tocorporatism, the state may have bilateral relations with employers (e.g. variousperiods of incomes policy in France) and also shape employers’ conduct throughlegally mandated collective bargaining. Finally, the state can play a critical rolein the character of market and managerial relations. In Anglo-Saxon countries,the two have not been distinguished, and a collective agreement may coverseniority rules and discipline as well as wages and conditions. The sharpestcontrast is Germany, where unions handle wages and conditions and have theright to strike on these matters but where legally mandated works councils dealwith a range of other issues, including work organization and stafﬁng and dis-ciplinary questions, but do not have the right to strike. Many other countrieshave collective structures in addition to trade unions. As discussed in chapter 6,the issue of legally underpinned rights of information and consultation hasemerged in the UK, and is likely to grow in signiﬁcance.Conﬂict, power and frames of referenceAn understanding of the nature of workplace rules can be developed by con-sidering three perspectives on rules, usually termed ‘frames of reference’. Theorigin of the debate on frames of reference was a distinction made by Fox (1966)between unitary and pluralist approaches.• The unitary view is that there is an identity of interest between employer and employee. Any conﬂict that may occur is then seen as ‘the result of mis- understanding or mischief; in other words, as pathological’ (Crouch 1982: 18). This view underlay much taken-for-granted managerial thinking about everyone in an enterprise having shared goals, and also underpinned several academic approaches, notably the ‘human relations’ tradition (see Rose 1988). Unitarism was often used as a straw man representing old-fashioned and unrealistic ideas, but surveys found that many managers continued to believe in a harmony of interest, and, as should already be clear, a resurgence of managerial self-conﬁdence and a reassertion of market individualism under- pinned a revival of unitarism from the 1980s. During the 1990s, HRM often implied that management was the sole or at least key authority. HRM prac- tice is likely to have a strong unitary aspect, as reﬂected in the ﬁnding of the
THE EMPLOYMENT RELATIONSHIP 11 1998 Workplace Employee Relations Survey that 72 per cent of workplace managers responsible for personnel matters prefer to consult directly with employees rather than with trade unions (Cully et al. 1999: 88). Managers without these responsibilities are likely to be even more strongly ‘unitarist’.• Pluralists see conﬂict as inevitable because, to cite Clegg (1979: 1), various organizations participate in determining the rules of employment. These have their own bases of authority, and ‘whenever there are separate sources of authority there is the risk of conﬂict’. Pluralism underlay the views of the Donovan Commission, which was established in 1965 to analyse the increas- ingly conﬂictual state of industrial relations and whose analysis encapsulated some basic assumptions, notably the reassertion of the value of voluntarism. Pluralism was particularly salient in the approach of management: instead of a unitary denial that there was any rational basis for conﬂict, managers should recognize the inevitability of disputes and seek means to regulate them. In Flanders’s (1970: 172) oft-quoted dictum, ‘the paradox, whose truth managements have found it so difﬁcult to accept, is that they can only regain control by sharing it’.• A third, radical, approach developed as a critique, or in the signiﬁcant case of Fox (1974) an auto-critique, of pluralism (see also Hyman 1978; Edwards 1998). Pluralists assumed, ﬁrst, that reform could be in the interests of all, thus neglecting major differences of interest between workers and managers, and, second, that institutional tinkering could meet the goals of a reformist management, thus failing to acknowledge that ‘disorder’ ran much deeper than a weakness of institutions. Much of the academic debate on these approaches treated them as mutuallyexclusive and incommensurable. Each approach also bore the mark of its origins:unitarism in human relations traditions, pluralism in organized collective bargain-ing, and radicalism in shop-ﬂoor discontent that seemed immune to all attemptsat institutionalization. Yet it would be as wrong to write off radicalism on thegrounds of the apparent disappearance of this discontent as it was to see unitarismas simply naive and outdated. A biography of radicalism’s key inﬂuence, KarlMarx, notes that an investment banker told New Yorker magazine in 1997 that‘Marx’s approach is the best way to look at capitalism’ (Wheen 1999: 5). Eachapproach has some strengths, though an appropriately explicated radical view isin my view analytically the best means to understand the nature of the employ-ment relationship (Edwards 1998). Consider the unitary view. To assume that all conﬂict is pathological is plainlyan unsatisfactory view of organizational life. Yet the view made two key points.First, surveys have found that managers, and indeed many workers, tend to seetheir ﬁrms in unitary terms; for example, when asked whether a ﬁrm is like afootball team or whether employers and workers are on opposite sides, workersoften choose the former. The ﬁrst UK study to ask this question found that 67per cent of a sample of manual workers agreed with the statement that ‘team-work means success and is to everyone’s advantage’ (Goldthorpe et al. 1968:73). Similarly, overt disputation is relatively rare.
12 PAUL EDWARDS Second, at the analytical level there are areas of shared interest: if workersand managers were totally opposed to each other, workplace relations wouldsimply break down. Consider some deeper analysis of the football team analogy.In reviewing the team analogy, Ramsay (1975) noted, ﬁrst, that Goldthorpeet al. did not equate teamwork with harmony but rather ‘interdependence’ ofmanagement and worker: teamwork was likely to mean pragmatic acceptance ofthe need for co-operation and not a completely shared vision with management.Second, when workers have been asked whether teamwork speciﬁcally describestheir own situation, the proportion saying that it does declines. Third, in his ownwork Ramsay asked workers whether they agreed with the team view ‘becausepeople have to work together to get things done’ or because ‘managers and men[sic] have the same interests in everything that matters’; respondents split aboutsix to one in favour of the former. Pragmatic acceptance of current conditionsand not ideological agreement with management was predominant. As this bookshows, such results continue to have resonance, in particular in relation to HRMand commitment (see chapters 13 and 14). As for the pluralist and radical views, there may at one time have been cleardistinctions. Differences certainly remain, but the debate has moved on. Britishpluralism proved to be ﬂexible. Clegg (1979) responded to radicalism in a meas-ured way, arguing that pluralism could embrace many of the radicals’ pointsand that for many practical purposes there was nothing to choose between theperspectives. This contrasts with the situation in the United States where con-ventional writers (Kochan 1982) simply dismissed the radical critique (Hyman1982). Though this difference is hard to explain, one reason is surely the open-ness of British pluralism (see Edwards 1995). In particular, the stress on theinevitability of conﬂict at the point of production was compatible with a pluralistview. In his pluralist phase Fox (1966: 14) had noted that ‘co-operation needsto be engineered’; that is, securing workers’ consent is an active and uncertainprocess. Flanders (1964: 243–4) had earlier drawn on the important work ofBaldamus (1961) to argue that bargaining was a continuous and uncertain pro-cess. This is not to say that pluralism and radicalism are identical, but therehas been constructive debate, out of which the approach developed below hasemerged. Finally, note the word ‘power’ in the heading of this sub-section. An allegedfailing of IR texts is their lack of attention to this concept (Kelly 1998: 9–12). Itis true that explicit discussion is often absent. Authors of the classic texts, such asClegg, might well have replied that power and conﬂict are the very stuff ofindustrial relations, for the negotiation of rules necessarily entails power andinﬂuence, and hence that separate discussion was redundant. They might alsohave said that it was the place of disciplines such as political science to debatethe concept, and that IR could use the results. There are of course problems withsuch a neat division of labour, but it has a point. IR is a ﬁeld of study, andcannot debate the fundamentals of concepts developed in politics, sociology,economics, and psychology. Yet it does have an underlying view of power whichmight be summarized in propositions such as the following.
THE EMPLOYMENT RELATIONSHIP 13• Power is a capacity to pursue one’s own interests, and it can be activated through individual or collective means.• Power involves the capacity to oppose the actions of others (reactive power) and to pursue one’s own objectives (proactive power). For example, in the UK, trade unions are often seen as having had reactive power but as lacking the means to pursue a proactive agenda.• As in other areas of social life, power is embedded in continuing relation- ships, and establishing ‘interests’ is never easy. For example, if a manage- ment proposes a new payment scheme, workers may favour it on the grounds that it is ‘fairer’ than a previous scheme and that it offers the prospect of increased earnings, but have doubts as to the chance that the promise will be realized and perhaps also fear managerial good intentions (e.g. does the scheme foreshadow a move towards individual merit pay, or job losses or . . . ?)• Power resources can shift over time (most obviously the declining power of trade unions since 1980).• Resources are not ﬁxed ‘things’ but are also developed through use. For example, it is often found that new forms of work organization ‘fail’ for lack of managerial commitment (i.e. the resources were not in fact deployed effectively) or because the initiatives run counter to other activities (i.e. employment relations have many aspects, and the power to impose a new work organization may run counter to the need to retain employee consent).• Power resides in organizational routines and assumptions as well as in overt actions. Managements may exert power over workers by shaping expecta- tions, but workers also have resources which they can mobilize, so that power relations are necessarily ﬂuid and uncertain. Such themes run through this book, and the reader may ﬁnd it helpful to bearthem in mind. Many IR studies, of which the work of Armstrong et al. (1981)still stands out, also clearly deploy analyses of power. They use concepts dis-cussed further below which reﬁne the above bullet points.Rules, power and the negotiation of orderIt is useful to begin with the nature of rules. Rules do not have to be clearlyenunciated, and many of the most important ones are not. A long series ofshop-ﬂoor studies (summarized in Edwards 1988, 2000) has revealed that ex-pectations about how work is to be performed often arise from informalunderstandings. For example, a worker new to an establishment may discoverthat a supervisor permits workers to leave early at the end of a shift. She maythen learn that this concession is granted only when work is slack or when astrict manager is absent, and that it is not wise to advertise it too widely. Shemay even ﬁnd that this local understanding counts for nothing if managersdecide to enforce the formal rules. Whether or not managers in fact enforce theformal rules and how they do so will depend on a variety of factors. Theseinclude:
14 PAUL EDWARDS• The procedures of the ﬁrm. If it has a system of warnings and appeals, it may issue a warning for a ﬁrst offence, but in the absence of such disciplinary procedures it may act in a more sudden and less predictable way.• The presence of a union. Can a representative make out a case for clemency on the grounds of the inexperience of the worker concerned or that the relevant practice had become taken for granted?• The role of the law. If a worker is dismissed for a breach of a rule, can it be shown that the dismissal was fair in the circumstances? This example shows that ‘the rules’ are many and varied, that different typesof rule may apply to any given situation, and that rules have to be interpretedin action for them to have any practical meaning. The status of a rule also varies.A loose understanding may indicate normally accepted practice. But it may havelittle force. When understandings attain rather more acceptance and legitimacythey may be termed custom and practice rules. As the classic study of the subject(Brown 1973) shows, managements may unwittingly allow one-off concessionsto grow into established expectations. Where workers have the power to insistthat the expectations are honoured, a custom and practice rule is born. A laterstudy showed that managers, too, generate custom and practice rules (Armstrongand Goodman 1979). In one case, a written rule in a collective agreementrequiring that workers be given notice if they were to be laid off was successfullyignored by managers who pointed out that workers who stood on their rightswould be entitled to only their low basic rate of pay, whereas if they went homeearly they could ‘get a lift with the housework’. Managers here used power topersuade workers where their own interests lay. Finally, why does not custom and practice continue to grow by a process ofaccretion? One important answer is that managements crack down on activitieswhich get out of hand. They may do so on a piecemeal basis (for example inmuch of the UK car industry during the 1950s, when managements wouldattack shop-ﬂoor leaders when immediate conditions allowed, but withoutrooting out the challenge and in fact helping further to embed it: Jefferys 1988)or as part of a general campaign. Examples of such campaigns became familiarfrom the 1980s as managements reasserted their authority and rooted out for-merly tolerated practices. The point is particularly signiﬁcant in Britain. The lackof legal enforceability of collective agreements, combined with the preferencesamong managements and unions for informality, means that settling issuesthrough unwritten understandings has played a particularly large part in theway in which the rules of employment are generated and sustained. Why is the making of rules so difﬁcult? A key reason is that the employmentcontract is indeterminate. In a commercial contract, a product or service issupplied for a price. In the labour contract, the worker sells an ability to work,which is translated into actual labour only during the course of the working day.Expectations about standards of performance have to be built up during theprocess of production. A rule is a complex social institution, not just a fewsentences in a rule book. It can comprise beliefs, ideologies and taken-for-grantedassumptions as well as formal provisions of rights and obligations. As noted
THE EMPLOYMENT RELATIONSHIP 15above, the actual operation of legal rights in the workplace depends on thepower, knowledge and organization of the parties as well as on the statute book.Perspectives on rulesThese points remain as valid today as they were when workplace custom andpractice was relatively well established. The current concept of the ‘psychologicalcontract’ tries to capture the idea of explicit and implicit expectations amongemployees about what their work will deliver. The concrete functioning ofinformal negotiation has changed, but the analytical principles remain import-ant, as seen, for example, in explanations of why changing IR institutions provesharder than might seem at ﬁrst sight (see e.g. chapters 7, 11 and 12). Threelevels of analysis may be distinguished. The ﬁrst level concerns the immediate balance of co-operation and conﬂict. Rad-icals and labour process writers, for example, allegedly saw everything in termsof conﬂict and managerial efforts to control workers more completely. Yet noserious discussion would deny that there can be shared interests (for example,workers may develop new abilities when advanced technology is introduced,as well as beneﬁt from the employment security of working for a successfulﬁrm), while also recognizing potential lines of tension (the technology mayplace new demands on workers and reduce the scope for informal control ofthe pacing and timing of work effort). The point is not whether employers andworkers have interests that are shared or that conﬂict. It is how these dimen-sions of the employment relationship are organized: how far does new techno-logy, for example, promote both the shared interest of working for an advancedcompany and possibly conﬂicting interests around the work practices that it mayentail? Second, the broader policies underlying workplace relations received attention. Tocontinue with the example of new technology, what does it mean to say thatit entails certain work practices? Are these determined by the technology or, asmany writers began to argue, the product of managerial choice? Although at thelevel of the individual workplace certain developments may seem inevitable,seen more broadly they may themselves reﬂect choice. There are two aspects ofanalysis here.• The ﬁrst considered the various approaches to labour regulation that man- agers might pursue, with the concept of managerial strategy being intens- ively debated. From a relatively orthodox IR position, Purcell and Sisson (1983) identiﬁed a set of ‘styles’. The way in which analysis was developing is illustrated by their inclusion of two styles which lay outside the usual IR focus on collective bargaining. These styles were an authoritarian non-union approach and a ‘sophisticated paternalist’ style, the latter generally involving a refusal to recognize unions and the intensive fostering of a sense of com- mitment to the company. Seen in retrospect, these styles foreshadowed what were later seen as two leading patterns of the management of labour, respect- ively, cost minimization and ‘high-commitment’ policies.
16 PAUL EDWARDS From a labour process approach, Friedman (1977) introduced the distinc- tion between the strategies of responsible autonomy and direct control. These strategies can be seen as underlying the more concrete styles identiﬁed by Purcell and Sisson. For Friedman, managements faced the problem of secur- ing workers’ co-operation while controlling them so that they would con- tinue to accept the authority of management and to work as directed. The two strategies represented approaches based on one or the other approach to the problem. Though Friedman sometimes presented the strategies as polar opposites, it is preferable to see them as elements which can be combined in various ways. Thus an employer may introduce quality circles to try to release workers’ creativity while also asserting ‘direct control’ over issues such as absenteeism and time-keeping. The analytical task is to show how the various strands of labour manage- ment are connected. For example, a major theme to emerge in Britain con- cerned the lack of deliberate linkages and the absence of a coherent approach implied by the term ‘strategy’. A theoretical perspective on this was provided by Hyman (1987), who argued that, because ﬁrms pursue the contradictory objectives of consent and control and because, moreover, they are operating in an unpredictable external environment, strategies must be routes to partial failure. That is, a strategy is not a neat package producing clear outcomes but necessarily contains several competing elements and has to be constantly reinterpreted as new results emerge and as the world changes. Management, in short, is not only a continuous, active and uncertain process but also neces- sarily involves the balancing of forces which are pushing in opposing directions.• The second aspect of analysis concerned the environment of labour man- agement policies. The links between the regulation of labour and business structure and strategy received considerable attention. How far are different approaches to labour the product of different product market circumstances? For example, does a competitive situation promote certain approaches and retard others? Attention was also directed not at variations between ﬁrms but at the overall environment in which they operated. How far is the labour policy of British ﬁrms shaped by the macroeconomic circumstances of the country and by generic features of its operation, notably the education and training of the workforce? The third level of analysis concerns the fundamental nature of the employmentrelationship. Many texts note that conﬂict and co-operation are both important,but they tend to stop at this point. This raises the question of whether conﬂictis any more than an occasional accident and whether it is more basic thanco-operation. The key point about the indeterminacy of the labour contract andstrategies of labour control is that managers and workers are locked into arelationship that is contradictory and antagonistic. It is contradictory not in thesense of logical incompatibility but because managements have to pursuethe objectives of control and releasing creativity, both of which are inherent inthe relationship with workers and which call for different approaches. The rela-tionship is antagonistic because managerial strategies are about the deployment
THE EMPLOYMENT RELATIONSHIP 17of workers’ labour power in ways which permit the generation of a surplus.Workers are the only people who produce a surplus in the production pro-cess but, unlike the independent artisan, they do not determine how their labourpower is deployed to meet the objective. There is thus a relation of ‘structured antagonism’ between employer andworker (Edwards 1986). This term is used to stress that the antagonism is builtinto the basis of the relationship, even though on a day-to-day level co-operationis also important. It is important to distinguish this idea from the more usualone of a conﬂict of interest. The latter has the problem of implying that thereal or fundamental interests of capital and labour are opposed, and hence thatany form of participation scheme is simply a new way of bending workers tocapital’s demands. The fact that workers have several interests confounds thisidea. A structured antagonism is a basic aspect of the employment relationshipwhich shapes how day-to-day relations are handled but is not something whichfeeds directly into the interests of the parties. Firms have to ﬁnd ways to con-tinue to extract a surplus, and if they do not then both they and their workerswill suffer. Balancing the needs of controlling workers and securing commit-ment rests ultimately on ensuring that a surplus continues to be generated. Itmay well be in workers’ interests that it is indeed generated, but this should notdisguise the fact that they are exploited. The contemporary signiﬁcance is simply that much workplace change is pres-ented as though it cuts through old relations of conﬂict to promote total unity.Yet any unity has to be actively created, and it cannot be total because of thestructural conditions in which employers and workers ﬁnd themselves.Methods of enquiryWhat methods have been used to pursue this agenda? A feature of many oldertexts, and indeed some more recent ones, is the detailed account of institutionsand of how bargaining is carried out. There is little self-conscious discussion ofmethodology or of exactly how information is gathered. This reﬂects the way inwhich they are written: they draw on their authors’ personal knowledge of theoperation of procedures, which is generally backed up by statistics on wages orstrikes and the results of ofﬁcial inquiries. Over the past two decades, analysishas become less institutionally focused and more conscious of the nature of theresearch base. Two developments stand out. First, a series of surveys has been conducted. The best known are the threeWorkplace Industrial Relations Surveys (WIRS) of 1980, 1984 and 1990 andtheir successor of 1998, now called the Workplace Employment Relations Sur-vey (WERS). These are based on large samples which are representative of thegreat majority of workplaces in Great Britain; each survey has involved interviewswith the manager responsible for IR and where relevant an employee repres-entative, and in 1998 there was also a survey of employees. The surveys notonly provide a mass of information about institutional arrangements such as thestructure of the personnel function and the extent and coverage of collectivebargaining. They also contain important information on working practices such
18 PAUL EDWARDSas the extent of systems of communication and involvement and the use ofteamwork. The survey design has included a panel element which has beencrucial in distinguishing two possible sources of change. For example, coverageof collective bargaining in 1998 was much lower than it was in 1980. This couldbe due to one of three sorts of change:• changes within continuing establishments, that is, a reduced propensity to engage in collective bargaining;• structural changes, whereby sectors where bargaining is most common consti- tute a declining proportion of the population of establishments; and• changes due to ‘births and deaths’, so that even within a given sector newly established workplaces are less likely to engage in collective bargaining, an effect which may be further distinguished according to whether the ‘births’ differ from the ‘deaths’, or from deaths and also continuing establishments. Such differences are considered throughout this volume, but two contrastingﬁndings may be highlighted now to illustrate the potentially different dynamics(Millward et al. 2000: 105, 190):• Change in the presence of trade unions in the workplace has reﬂected the third inﬂuence: continuing workplaces were stable in the presence of unions between 1990 and 1998, but newly established workplaces were much less likely to have a union presence.• By contrast, there was a shift away from multi-employer collective bargaining even among continuing workplaces, so that in this respect the ﬁrst inﬂuence was operating: the same managements were changing their policies. Second, case-study work has explored the processes underlying the patternsdescribed by surveys. Case studies have developed in three main ways. First,their theoretical orientation has developed in the light of interests in the labourprocess and management strategy. Second, though it is sometimes said that therise of HRM led to a neglect of the worker’s point of view, in fact many studies(e.g. those discussed in chapters 13 and 18) have addressed the effects of changeon workers, thus developing the point that the rules of employment are not justabout the activities of managements and trade unions. Third, some researchershave used case studies in more than one location, in order to explore variationsin behaviour and thus deepen the explanations offered. IR research combinessurvey and case-study methods, as well as using established methods such asanalysis of ofﬁcial statistics. The following chapters draw on these methods invarying degrees and where relevant they highlight gaps in knowledge as well asdifferences in interpretation.Contract and statusWork relationships in the UK, and indeed in most advanced industrial econo-mies, have, then, seen many confusing and apparently contradictory trends. One
THE EMPLOYMENT RELATIONSHIP 19device to understand these trends is the distinction between contract and status(Streeck 1987). ‘Contract’ refers to a relationship based on the principles of ‘hireand ﬁre’ and individual responsibility. It is thus related to the more obvious ideaof a relationship based on pure market principles, but is also broader in recogniz-ing that speciﬁc and detailed obligations can be provided through organizationalrules and not the market. For example, many organizations use systems ofperformance appraisal which are deﬁned by the organizations’ rules rather thana market in any exact sense of the term. ‘Status’ covers longer-term relation-ships and the treatment of the employee as an investment rather than a cost;for current purposes, we may include the provision of means for employees toparticipate in decision-making. A purely contractual relationship will give nospace for such ‘voice’ and will treat employees simply as hired help, whereasin a more lasting relationship members of an organization may be expected toengage in its key choices. There are two reasons, which run through manydebates on the employment relationship. The ﬁrst relates to efﬁciency: givingworkers a say may improve organizational functioning in some way. The secondrelates to equity: employees investing in a long-term relationship have rightswhich need to be respected. Europeanization would imply an emphasis on status while Americanizationimplies more stress on contract. Some trends in the direction of contract andstatus are more securely documented than others, as this book aims to show. Forthe present, we simply note the following. Trends towards contract would include:1 A rise in the use of temporary and agency workers. The increase is, however, from a low base, and quite small proportions of the workforce are involved.2 An increased reliance on the measurement of performance and the tying of reward to this measurement. Performance measurement is certainly extens- ive, but direct links to reward are highly variable, and many appraisal schemes are very standardized rather than offering a strictly individual market ex- change between each worker and the employer.3 A decline in the role and inﬂuence of trade unions, which were traditionally the only real means for UK employees to express their voice collectively (works councils – representative structures not based on trade unions but instead being elected by all employees of a ﬁrm – on the continental Euro- pean model being largely absent). Trade union membership fell rapidly during the 1980s and 1990s, with little replacement by other representative structures; hence the widespread concern about a ‘representation gap’.4 Insecurity in the sense of (a) an increased objective chance of job loss and (b) an increased perception that job prospects are uncertain. There is evidence of insecurity. One measure is job tenure where, as noted above, there have been contrasting trends for men and women. Staying in a job can, however, represent insecurity because of fear that another job will not be found. Other aspects of insecurity include career insecurity: a concern that a predictable, or at least manageable, progression of jobs is not possible. Heery and Salmon (2000) argue that the combination of increased risks of unemployment, a
20 PAUL EDWARDS rise in temporary and casual work, a reduction in legal protections, and a rise in employee perceptions of an uncertain future together point to a growth of insecurity, though they also note that there will be important differences between sections of the workforce. In relation to status, trends in employee voice running counter to that justmentioned include an increase in the use of direct rather than representativeparticipation, for example the use of teamworking and problem-solving groups.Status would also be reﬂected in other practices associated with human resourcemanagement, such as improved communication with employees, a move towards‘single status’ (that is, treating all employees the same in terms of such things asworking hours, sick pay and pensions, in place of long-entrenched distinctionsbetween non-manual and manual workers), and the use of proﬁt-related payand share-ownership schemes. There is evidence of growing use of many ofthese measures, though it is also well established that the adoption of them inso-called bundles of high-involvement practices remains very rare. Streeck’s summary of moves towards contract and status is given in table 1.3,which also indicates where the matters are pursued in this book. This listinganticipated several developments, notably around teamwork and shifts awayfrom collective bargaining, which have become more salient since 1987. A par-ticularly foresightful point was the highlighting of ‘the possible contradictionsinherent in a simultaneous pursuit of restored contract and extended status’ and thefact that employers will ‘ﬁnd it exceedingly hard to formulate a consistentstrategic approach to building a more ﬂexible system’ of employment relations(Streeck 1987: 295, emphasis added; see further chapter 7 on the differentdimensions of ﬂexibility). As discussed below, contract and status are not poleson a continuum but separate dimensions which can vary independently.Table 1.3 Two routes to organizational ﬂexibility Return to contract Extension of status ChapterEmployment status of workers Temporary Permanent 3Numerical ﬂexibility Hire and ﬁre Flexible working 7, 14 timeFunctional ﬂexibility Hire and ﬁre Self-regulated 13, 14 job rotationWork organization Taylorist Teamwork 13, 14Qualiﬁcations sought Narrow Broad 15Wage determination Industrial Payment by 8, 12 engineering abilityManagement style Unilateral Consultation, 7, 9, 10, prerogative participation 12Source: Abbreviated from Streeck (1987: 294).
THE EMPLOYMENT RELATIONSHIP 21 One recent research study illustrates the problems of developing consistentemployment relations strategies. Grimshaw et al. (2001) draw on studies of fourorganizations to demonstrate shifts away from structured approaches associatedwith well-ordered internal labour markets (systems with clear promotion laddersand established pay structures). They argue that they have been replaced witha contradictory mix of policies. As an earlier study (Marginson et al. 1994) putit, employer policies were increasingly ‘eclectic’. The implication is that we areseeing complex efforts to use aspects of contract and status rather than a coher-ent shift from one form of employment relationship to another. Several strandsrun through this book’s efforts to make sense of these developments.• First, contract and status can obviously both exist if they apply to different groups of workers. For example, temporary workers may be subject to con- tractual relations while permanent ones enjoy status. There is also some evidence of a polarization of experience, for example rising skill levels in some but not all occupations, so that contract and status receive different emphases in different jobs.• Second, they often coexist in the same workplace, for example where em- ployees work in teams but are also subject to closer performance measure- ment. Figure 1.2 attempts to make this point more analytically. Note ﬁrst the dimensions. If we classiﬁed status and contract according to their presence or absence, we would have one area of the diagram where neither was being used. It probably makes more sense to think in terms of how developed and explicit the adoption of one dimension or the other is. The conventional cases are in the bottom right and top left. The ﬁrst of these is a traditional ‘hire and ﬁre’ employment relationship, but note that for contract to be developed this will need to entail formal and explicit manage- rial rules specifying the terms of the contract and the consequences of a breach. High status and low contract is associated with what is now termed ‘soft HRM’, meaning an emphasis on training, career development and ‘em- powerment’, but it would also characterize much more traditional paternalist STATUS Strong Paternalism Performance management ‘Soft’ HRM Arbitrary Hire and ﬁre, management power formal discipline Weak CONTRACT Weak StrongFigure 1.2 Status and contract
22 PAUL EDWARDS relationships based on unwritten understandings and diffuse obligations. Now, research literature on both these situations ﬁnds that in practice there is often a hard edge to the relevant relationships. The point is not to say that, empirically, situations lacking hard and coercive elements are at all likely. It is to establish that there can be situations where there are tendencies in this direction, and to indicate a benchmark against which actual practice can be measured. The case in the top right is of particular interest. It identiﬁes cases where status and contract are both emphasized, notably through such practices as performance management, where employees may have a degree of autonomy and even ‘empowerment’ but are also held to account against explicit, hard performance standards. Finally, the reverse position is also important: status is not important, and contract is weak in that there are no formal and explicit rules: the employment relationship is weakly institutionalized. For long periods before the arrival of trade unions, this model applied to large parts of the workforce. Trade union and legal and other pressures have moved many employers away from the extremes of this, but the model still characterizes some employers and it also captures a tendency that can be found widely, for example the desire to ‘get rid of problem staff’ quickly and the consequent tendency to ignore formal procedures. Chapters 7 and 12 focus speciﬁcally on employer strategies. Chapter 14 illustrates such themes through case studies of two organizations which might be expected to have clear strategies but which in practice combine different ways of managing labour in complex ways. Eclecticism remains a central feature of how the tensions of the employment relationship are handled.• Third, it is impossible to add up different dimensions to produce a simple overall balance sheet. For example, the decline of unions has entailed a signi- ﬁcant reduction in the extent to which many key issues of pay and working conditions are subject to any serious discussion with employees or their representatives. It is not possible to say that this is ‘balanced’ by any growth in direct participation, for the dimensions are different.• Fourth, what benchmark is used in assessing the state of industrial relations? One is an alleged past of powerful trade unions and perennial strikes, as represented in the continuing use of the image of the ‘winter of discontent’ (a period of industrial militancy in 1978–9, the label apparently being intro- duced, so the IR academic W. E. J. McCarthy has discovered, by an inspired sub-editor on the Sun). Yet trade unions covered at their peak only half the workforce; strikes were concentrated in a very narrow sector of the economy and even there were quite rare; and, as in any conﬂict, militancy takes two sides, so that to jump from the existence of many strikes to the presumption that the ‘cause’ was militant and irresponsible unions was a common but plainly faulty leap of logic. Another model uses heady labels such as ‘employee empowerment’, the ‘skills revolution’ and the ‘learning society’. As this book will show, there has been change in relevant areas, but there remains a large gap between the rhetoric of these terms and the reality of the limited extent and depth of change. IR research has a strong sense of history (see chapters
THE EMPLOYMENT RELATIONSHIP 23 2 and 14), and thus helps to counteract the many claims that work is funda- mentally different from its shape in ‘the past’.The Changing Character of Industrial RelationsThe above description reﬂects IR scholarship going back many years. It treatsIR as a process largely conﬁned to a nation-state and has thus required amend-ment in the light of supra-national regulation and ‘globalization’. There havealso been challenges to the intellectual integrity of IR as a ﬁeld of study. Theyarise from the emergence of HRM (which offers a different way of looking atestablished themes in the management of labour) and some broader analyses(which suggest that these themes need to placed in a wider context).Industrial relations systems and globalizationAs noted above, writers on industrial relations systems have tended to presentthese systems as self-contained. Inﬂuences and borrowings were recognized,for example the role of the British in shaping the German post-war ‘system’, theAmerican inﬂuence on Japanese post-war reconstruction, and direct borrowingsby the British Conservative government of 1970–4 of aspects of American in-dustrial relations. Yet such events were seen as unusual and often unsuccessful.In the attention to nationally bounded ‘systems’ IR writers were far from alone.Indeed, it was well after Dunlop (1958) popularized the idea in IR that in thestudy of management more generally the concept of the national business system(that is, a distinctive and connected set of institutions of corporate governance)emerged (e.g. Lane 1989). Given their emphasis on uncertainty and the management of compromises,IR writers might have been more open than they were to challenges to the ideaof national systems. In the event, however, it has taken real-world events toundermine the idea. Three may be highlighted. First, several countries experi-enced a wave of Japanese direct investment which, though small in absoluteterms (see chapter 4), was seen as signiﬁcant in importing new ideas. The term‘Japanization’ was coined in 1986 (Turnbull 1986). The new ideas included closeattention to detail, a focus on the quality of goods and services, and continuousimprovement. Second, European legislation had an increasingly direct impactacross the European Union. During the 1990s directives on such issues asworking time and European Works Councils applied in the UK, and in additionmatters as diverse as pension payments and procedures for consultation aboutredundancy were subject to European legal decision. Third, the concept ofglobalization attained increasing prominence. Its precise meaning is disputed, asis the extent of the development which it characterizes, but the essential idea isthat economic activity takes on a global character, so that national systems losetheir distinctiveness and are increasingly inﬂuenced by international forces. Somewriters identify a ‘hollowing out’ of the nation-state, meaning that its inﬂuenceis weakened through the simultaneous rise of global regulation as in the World
24 PAUL EDWARDSTrade Organization, European monetary union, and the international transfer ofmodels of work organization by multinationals (see chapter 4). Some of the more popular ideas about globalization are ‘historicist’, simplyidentifying a trend and projecting it forward as well as, characteristically, por-traying the present as a complete break with the past and seeing the trend as anasocial and inevitable development (e.g. Ohmae 1995) that is the identiﬁcationof a trend and its projection into the future. (Note in passing the irony that histor-icism was seen originally, and most famously by the philosopher Popper (1957;see also Goldthorpe 1971), as a besetting failing of Marxism, and yet it is nowdedicated anti-Marxists who fall into the trap). Three key problems here are:• the failure to specify the causal processes underlying an observed trend, which is simply projected into the future;• the assumption that a trend in some parts of the economy will become general: it may be that portfolio careers can be identiﬁed where they did not exist, but they may reﬂect very speciﬁc circumstances;• the neglect of countervailing tendencies. These ideas are essentially positive about the effects of globalization. Othersuse the same method to portray an international ‘race to the bottom’ whereincapital seeks the cheapest site of production and thus undermines legal protec-tions to employees elsewhere. Such ideas have been popular since at least the1970s when the New International Division of Labour thesis made this argu-ment. Yet it was soon shown that the great bulk of world trade remained withinthe developed economies and that multinational companies generally retainedstrong bases in their countries of origin (see Hirst and Thompson 1996). More serious analysis speaks of competing models of capitalism and theirintersection: globalization is not a force of nature but an actively managed pro-cess (Coates 2000). For present purposes, we do not need to debate the theory ofglobalization, but may simply note some implications for IR in the UK. First, it isreasonably clear that the economy has been increasingly exposed to competitivepressures (whether these are labelled global or international is not important).Examples from the car industry were given above. Second, it does not followthat the coercive comparisons practised by large ﬁrms will necessarily leaddirectly to a shift towards countries with the lowest unit costs: there are manyreasons to produce in particular markets, and many considerations in locationdecisions. Third, the UK has beneﬁts as a production location, including relat-ively low wage costs by the standards of some European countries and, it isoften argued, labour laws which place few restrictions on redundancies andrestructuring. Fourth, issues of ‘best practice’ in the organization of work arelikely to have an increased salience, as are the links between IR and economicperformance. Fifth, however, many parts of the economy remain relatively freefrom direct international competition. Finally, there is little evidence of onedominant route to the management of employment relations. Different modelsremain in competition, and as this book shows American and European pressurescan both be discerned in the British context.
THE EMPLOYMENT RELATIONSHIP 25Industrial relations and human resource managementThe contrast between IR on the one hand and PM and HRM on the other maybe speciﬁed with the help of ﬁgure 1.2. The contrast with PM is conventionallymade in terms of the collective focus of IR and the individual focus of PM. Thereis a degree of truth in this, with IR having a strong emphasis on an organizedrelationship between managements and trade unions and with collective bar-gaining between them being both an empirical focus and a key analytical category.PM and HRM have given more attention to such ‘individual’ issues as recruit-ment and training. Until the 1980s, there was a tacit division of labour betweenIR and PM which was sometimes mirrored on the ground, with a ﬁrm’s indus-trial relations manager (often a man) dealing with collective bargaining andtrade unions while a female personnel or welfare ofﬁcer handled health andsafety, pensions, and the works canteen. This situation was thrown into ques-tion by two developments: a decline in the coverage of collective bargaining plusan even steeper decline in one of the staples of IR, organized industrial conﬂict,and a rise in HRM which claimed to offer an integrated approach to the manage-ment of labour, in which IR was only a small part. Expressions of HRM imperialism are, however, misplaced. Whether the rulesof employment are established collectively between an employer and its em-ployees acting in concert, or individually, is evidently a major empirical ques-tion, but IR is perfectly capable of dealing with both. And HRM has some majorlimitations of its own. Its starting point is how managers can organize their‘human resources’, which leads to two distinct issues. First, it takes the managerial perspective as its reference point: it asks, nothow employment is organized, but what managers can do to manage theiremployees. Employees are resources to be controlled and deployed, not equalparties to a relationship. Consider for example the features of HRM listed intable 1.4 that were identiﬁed by Storey (1992: 35) and are widely reproduced inHRM books. The approach is explicitly unitarist in its perspective, and it alsoseeks a move from collective to individualized relations with workers. It is truethat researchers in the ﬁeld are not necessarily committed to such a view. HRMTable 1.4 HRM and the management of labour Personnel and IR HRMNature of relations Pluralist UnitaristConﬂict Institutionalized De-emphasizedLabour management Collective bargaining Towards individual contracts contractsThrust of relations Regularized through Marginalizedwith stewards* facilities and training* i.e. shop stewards: elected trade union workplace representatives (see chapter 10 for details).Source: Abbreviated from Storey (1992: 35).
26 PAUL EDWARDSTable 1.5 Managerial approaches to labour management Strategic Non-strategicPeople as resource HRM PersonnelPeople as cost Not HRM? Traditional managementSource: Redman and Wilkinson (2001: 11). By permission of Pearson Education.is far from being a settled approach, and indeed one recent text starts by stress-ing its ‘diversity’ (Beardwell and Holden 1997). It is also true that many texts donot assume any simply harmony of interest between employer and employee;one even goes so far as to say that it adopts an ‘industrial relations approach’(Bach and Sisson 2000: 8). Yet such analytical views are consciously aimedat correcting a mismatch between the ‘rhetoric’ of harmony and the ‘reality’ ofconﬂicting interests. And the critique is needed because the ﬁeld starts froma managerial problematic even when it is not directly managerialist. An IRapproach starts from a different position. In the words of a leading US authority,IR ‘starts from an assumption that an enduring conﬂict of interests exists betweenworkers and employers’ (Kochan 1998: 37), though as argued above this is notin fact an assumption but a demonstrable characteristic. Second, as an analytical approach HRM tends to assume what it sets out toprove. Another recent book usefully brings out the point. Redman and Wilkinson(2001: 11) argue that people can be seen as a resource or a cost and that manag-erial approaches can be strategic or non-strategic, which generates table 1.5.HRM describes only one cell in the matrix and, as will be seen throughout thisbook, one which applies to only a minority of organizations. To be clear, it is notbeing argued that students of HRM are unaware of such points; indeed, themore analytical books provide a balanced and critical appraisal. But they draw,explicitly or implicitly, on IR writings to make sense of the terrain. For examplethe remarks quoted above by Bratton and Gold (1999: 28) on ‘tensions andcontradictions’ of HR strategies draw directly on IR analyses. Third, HRM gives little direct attention to the role of the state in the regulationof labour. It is more interested in policy within the enterprise and not thedynamic interplay between the state, employers, unions and employees. To see recent developments in the UK as a secular shift from a collective andconﬂict-based system to one based on individualism and consent is to commitseveral sorts of error. First, collective bargaining was never ubiquitous: at theirpeak unions covered only half the workforce, and organized overt conﬂict wasthe exception rather than the norm (see chapter 10). Second, the undoubteddecline in institutional industrial relations (declining union membership, manyfewer strikes, reduced union inﬂuence at national level) leaves open the issueof what has replaced it. As later chapters in this book show, a reasonable pic-ture is one of variation and fragmentation, with only a minority of employers
THE EMPLOYMENT RELATIONSHIP 27embracing anything like the full suite of HRM practices, and with many othersrelying on traditional approaches to labour based on cost minimization. Underlying these empirical issues are two theoretical problems. The ﬁrst iscommon in efforts to analyse contemporary developments, for example claimsthat ‘employability’ and ‘portfolio careers’ are replacing traditional bureaucraticcareer structures. They often fall into the error of historicism, In the IR sphere,the trend may be the decline of collectivism, but it would be rash to suppose thatthis will continue without limit or that it cannot be counteracted by forcesmoving in the opposite direction. The second theoretical issue turns on the conceptualization of the employ-ment relationship. IR research has been predicated on the assumption that therelationship is one of conﬂict, power and inequality. The core of IR is that it seesthe employment relationship as based on a structured antagonism, with anyunity and co-operation being built on uncertain foundations and with the cre-ation of consent being inherently partial and uncertain. An irony here is that proponents of new techniques such as business processre-engineering recognize in passing that co-operation indeed has to be engi-neered. Consider the following from an HRM text: management’s actions often have unintended consequences. Thus, for example, the[re is a] paradox contained in the prescriptive advice to managers which encour- ages leaders to ‘gain control by giving it up’ (Champy 1996). . . . [This paradox] suggests that because of persistent and fundamental continuities in the post- industrial labour process, there is no such thing as the ‘right’ human resource strategy, system or technique and that, whatever systems are adopted, they will have to be regularly modiﬁed or replaced as their internal tensions and contradic- tions appear. (Bratton and Gold 1999: 28)The irony is the rediscovery of Flanders’s language discussed above withoutreference to it, yet with the underlying unitarist assumption that re-engineeringis a process managed from the top in the interests of all. The idea that there arelegitimate competing sources of authority seems alien, as does Flanders’s discus-sion of the tendency of managers to rely on their right of command and theirfailure seriously to engage with employees (see discussion of contract and statusbelow). Business process re-engineers could read the conclusion of Flanders’saccount of an earlier effort at re-engineering with proﬁt (Flanders 1964).Challenges to IR as a domainIt was argued above that British pluralism responded constructively to some chal-lenges, notably the radical view of the negotiation of order on the shop ﬂoor. Ithas arguably been less open to other challenges. Three may be highlighted.• Worker interests. According to Kelly (1998), IR retains a descriptive approach, its research agenda is unduly driven by employers, and, as noted above, it lacks a proper theory of power. As a result, ‘we don’t know’, for example, ‘precisely why and how union power declined or by how much in the 1980s’ (1998: 23). The answer is mobilization theory and in turn ‘the fulcrum of the
28 PAUL EDWARDS model is interests and the ways in which people (particularly members of subordinate groups) come to deﬁne them. To what extent do they believe their interests to be similar to, different from, or opposed to, those of the ruling group?’ (Kelly 1998: 25, emphasis original). Other concepts address the degree to which a group organizes around given interests, mobilizes against other groups, and has the opportunity to pursue its interests. This theory has, says Kelly (1998: 126–9) three advantages over ‘rival approaches’: – instead of starting from the employer’s need for co-operation and to secure work performance, it starts from injustice and exploitation; – it does not depend on a simple distinction between individualism and collectivism, but distinguishes interest deﬁnition, organization, mobiliza- tion, and so on, treating ‘as problematic what previous industrial relations researchers often took for granted, namely the awareness by workers of a set of common interests opposed to those of the employer’; – it helps address key issues such as how employees deﬁne interests in particular ways.• Gender. Traditional IR, according to Wajcman (2000), took the male worker as the norm and saw women as marginal or of secondary interest. More fundamentally, it was not just that women as a group were neglected despite their substantial and growing place in the workforce. The gendered processes underlying much of the substance of IR were also neglected. The very institu- tions of IR are not gender-neutral. For example, payment systems contain gendered assumptions in how they deﬁne and measure the attributes of jobs, so that jobs typically performed by women will be rated differently from male jobs. Collective bargaining was not well attuned to the representation of equality agendas (Colling and Dickens 1998). Men’s work is also gendered, for example in the ways in which managers tend to recruit people in their own image, and an informal ‘men’s club’ atmosphere creates expectations as to acceptable styles of behaviour.• Work and society. Ackers (2002) criticizes IR for its focus on relations in the workplace, to the neglect of links with the family and other spheres, and for the absence of an explicit ethical dimension. He offers ‘neo-pluralism’ as the means to develop a critical analysis of management practice and an under- standing of the normative aspects of the regulation of employment. These views are plainly very different. Kelly focuses virtually exclusively onthe traditional terrain of worker–manager relations and has been criticized expli-citly by Ackers (2002) for the emphasis on economistic workplace militancy. Histext would not be viewed, I believe, as gender-sensitive. Kelly offers a deepeningof the agenda, that is a consideration of fundamental conﬂicts between workersand managers, rather than a broadening. Wajcman and Ackers aim to view theterrain in new ways and to link it to other aspects of society. These issues are raised here partly to stress that texts such as this have beenfound wanting in the ways speciﬁed, and to invite the reader to consider the
THE EMPLOYMENT RELATIONSHIP 29value of the criticisms during study of this book. It is not, however, whollysatisfactory to leave evaluation up to the reader, and some initial comments onthe three views are thus offered here. First, and in opposition to Kelly, in the identiﬁcation of a ﬁeld of study, afocus on employee interests is as unsatisfactory as one on those of employers.IR examines job regulation, but this is scarcely to adopt an employer’s agenda(whereas HRM arguably does adopt such an agenda). Second, Ackers reasonablypoints to the need to link IR to other aspects of society, which is not a new calland which raises the question of how broadly a ﬁeld is deﬁned. The key issue ishow far it makes sense to analyse a particular theme while leaving others to oneside. For example, the institutions of collective bargaining are shaped by mana-gerial policy elsewhere in the ﬁrm, and their operation in practice will dependon the negotiation of rules informally. In the same way, it can be argued thatchanges in the family, for example, may inﬂuence IR without necessarily need-ing analysis in their own right. The danger of broadening an approach is ofcourse that it loses all coherence. A useful model was introduced by Emmettand Morgan (1982; for further discussion see Edwards 1986: 280). In theiranalysis of workplace ethnographies, they liken the walls of a workplace to a‘semi-permeable membrane’, which ﬁlters inﬂuences from outside and also shapeshow processes within the workplace affect relations elsewhere. If the membranewere wholly permeable, then Ackers might well be right, though it might alsofollow that IR was simply the working out in the workplace of forces determinedelsewhere, which would make the subject of rather trivial importance. But, aswe have seen, and as for example studies of pay determination show (seechapter 8), processes of negotiation develop their own logic and dynamic. IR isof course inﬂuenced by other forces, and as a developing ﬁeld it needs to payattention to new ones, but it can be reasonably bounded as a ﬁeld with its ownissues. Third, Kelly’s arguments for mobilization are wholly consistent with theapproach of this book. To the extent that he directs attention to issues whichmay have been relatively neglected, he helps to re-emphasize some key themes.But, in many ways like this book, he retains a relatively orthodox view as towhat IR is about. Fourth, it is true that institutions and processes are not gender-neutral (andalso true that there are important lines of division other than gender). Yet thisdoes not mean that everything about them is so suffused with gender that theycannot be discussed without gender being included at every turn. More pre-cisely, we can identify at least four situations. Some institutions directly entailgender issues, for example equal opportunities policy and practice. Others arelikely to have less direct but still obvious gender implications, for example paystructures which are based on estimates of the value of certain skills (and whichhave been found to reward some (male) skills more than others). Others needhave no gender implications even though in practice they do so. For example,there is nothing inherent in trade union election procedures which excludeswomen and ethnic minorities, but the fact that unions have historically been male-dominated, together with their tendency to reproduce the gender assumptions
30 PAUL EDWARDSof society at large, has tended to produce marked gender differences in accessto positions of power. Finally, some institutions, for example those of cor-poratism, are relatively distant from gender divisions. It is thus reasonable tocontinue to address some institutions and processes without necessarily drawingout the gender implications. The reader may well wish to use these four categories to assess how well thisbook reﬂects gender issues. For example, under the second and third categories,how clearly are the gender implications drawn out? In doing so, some bench-marks may be helpful. The point of Wajcman’s critique is that IR research hasgiven insufﬁcient attention to gender issues; this text necessarily reﬂects thatresearch and cannot be expected to go far beyond it. It is also open to the readerto consider the implications of particular themes. For example, this book’s analy-sis of managerial strategies discusses such things as ‘ﬂexibility’, and plainly therecan be gender assumptions built into this notion. The analysis does not necessar-ily need to spell out the gender implications at every turn, and the key issue iswhether the tools of analysis can be deployed to address them. IR research isarguably reasonably well equipped to do this. Finally on this point, however, IRstructures and institutions have many implications, for class, ethnic inequality,and many other things. It is not reasonable to expect an IR book to deal withthem all, and nor should it: the argument here is that these structures andinstitutions deserve attention in their own terms, and they can be analysedaccordingly. Links with gender and other processes should not be neglected, but,to put the point bluntly, these links are only one part of a story which can betold without making them the central theme. Fifth, in relation to theory, some brief comments and references to socialtheory may be made, for those readers with a particular concern on this score.It is true that IR has always had a strong pragmatic bent, as is illustrated bythe saying (attributed to different members of the ‘Oxford school’ of reformist,pluralist scholars associated with the Donovan Report) that ‘an ounce of fact isworth a ton of theory’. It is also always useful to ask what we do not know andto question existing assumptions. Yet it would be wrong to infer that IR hasfailed to explain key issues within its domain. First, there are speciﬁc issueswithin the core domain of IR, such as the decline in union membership andin the role of union workplace representatives. Many such matters have beenextensively debated, and there is arguably a reasonable set of explanations.Second, there are more complex issues, such as the effects of new work organ-ization on employees. Plainly, one needs to distinguish different types of workerand different types of effect, and we are dealing with some effects such asperceptions of autonomy which are more complex than, say, the number ofpeople belonging to trade unions. Yet research has pursued an agenda whichis progressive in the sense that the topic under discussion is understood better(e.g. the balance between autonomy and responsibility) and the causes of differ-ent patterns are analysed. In like manner, the links between IR and variousoutcomes for workers and ﬁrms have been detailed. IR is less explicit about its theory than are the core disciplines of social science.Scepticism about the ability of a single discipline to explain complex phenomena